keyword-clusterer
关于
The keyword-clusterer skill groups keywords by semantic similarity and intent using embeddings, helping developers organize keyword lists into topic clusters. It's useful for content planning, site structure mapping, and identifying content gaps. This tool transforms raw keyword data into an organized content strategy framework.
快速安装
Claude Code
推荐npx skills add guia-matthieu/clawfu-skills -a claude-code/plugin add https://github.com/guia-matthieu/clawfu-skillsgit clone https://github.com/guia-matthieu/clawfu-skills.git ~/.claude/skills/keyword-clusterer在 Claude Code 中复制并粘贴此命令以安装该技能
技能文档
Keyword Clusterer
Group keywords by semantic similarity using embeddings - turn a keyword list into an organized content strategy.
When to Use This Skill
- Content planning - Group keywords into topic clusters
- Site structure - Map keywords to pages
- Intent analysis - Categorize by search intent
- Gap analysis - Find missing keyword themes
- PPC organization - Group keywords for ad groups
What Claude Does vs What You Decide
| Claude Does | You Decide |
|---|---|
| Structures analysis frameworks | Strategic priorities |
| Synthesizes market data | Competitive positioning |
| Identifies opportunities | Resource allocation |
| Creates strategic options | Final strategy selection |
| Suggests implementation approaches | Execution decisions |
Dependencies
pip install scikit-learn sentence-transformers pandas click
# For simpler usage without ML:
pip install click pandas
Commands
Cluster Keywords
python scripts/main.py cluster keywords.csv --n-clusters 10
python scripts/main.py cluster keywords.csv --column keyword --n-clusters 15
Find Similar
python scripts/main.py similar "content marketing" --count 20
Analyze Intent
python scripts/main.py intent keywords.csv --column keyword
Examples
Example 1: Cluster Keyword Research
# Input: keywords.csv with 500 keywords
python scripts/main.py cluster keywords.csv --n-clusters 12 --output clustered.csv
# Output:
# Cluster 1 (45 keywords): "content marketing"
# - content marketing strategy
# - content marketing tips
# - how to do content marketing
#
# Cluster 2 (38 keywords): "email marketing"
# - email marketing tools
# - best email marketing software
# - email campaign tips
# ...
Example 2: Categorize by Intent
python scripts/main.py intent keywords.csv --column keyword
# Output:
# Intent Analysis
# ──────────────────────
# Informational: 234 (47%)
# - how to, what is, guide, tips
# Commercial: 156 (31%)
# - best, top, review, compare
# Transactional: 78 (16%)
# - buy, price, discount, order
# Navigational: 32 (6%)
# - login, contact, brand names
Search Intent Categories
| Intent | Signals | Content Type |
|---|---|---|
| Informational | how, what, why, guide | Blog posts, guides |
| Commercial | best, top, review, vs | Comparisons, reviews |
| Transactional | buy, price, discount | Product pages |
| Navigational | [brand], login, contact | Landing pages |
Clustering Methods
| Method | Best For | Speed |
|---|---|---|
| semantic | Meaning-based grouping | Slower |
| lexical | Word overlap grouping | Faster |
| intent | Search intent categories | Fast |
Skill Boundaries
What This Skill Does Well
- Structuring strategic analysis
- Identifying market opportunities
- Creating strategic frameworks
- Synthesizing competitive data
What This Skill Cannot Do
- Replace market research
- Guarantee strategic success
- Know proprietary competitor info
- Make executive decisions
Related Skills
- content-repurposer - Create content for clusters
- lighthouse-audit - Optimize cluster pages
Skill Metadata
- Mode: centaur
category: seo-tools
subcategory: keyword-research
dependencies: [scikit-learn, sentence-transformers, pandas]
difficulty: intermediate
time_saved: 5+ hours/week
GitHub 仓库
相关推荐技能
content-collections
元Content Collections 是一个 TypeScript 优先的构建工具,可将本地 Markdown/MDX 文件转换为类型安全的数据集合。它专为构建博客、文档站和内容密集型 Vite+React 应用而设计,提供基于 Zod 的自动模式验证。该工具涵盖从 Vite 插件配置、MDX 编译到生产环境部署的完整工作流。
polymarket
元这个Claude Skill为开发者提供完整的Polymarket预测市场开发支持,涵盖API调用、交易执行和市场数据分析。关键特性包括实时WebSocket数据流,可监控实时交易、订单和市场动态。开发者可用它构建预测市场应用、实施交易策略并集成实时市场预测功能。
creating-opencode-plugins
元该Skill帮助开发者创建OpenCode插件,用于接入命令、文件、LSP等25+种事件。它提供了插件结构、事件API规范和JavaScript/TypeScript实现模式,适合需要拦截操作、扩展功能或自定义事件处理的场景。开发者可通过它快速构建响应式模块来增强OpenCode AI助手的能力。
sglang
元SGLang是一个专为LLM设计的高性能推理框架,特别适用于需要结构化输出的场景。它通过RadixAttention前缀缓存技术,在处理JSON、正则表达式、工具调用等具有重复前缀的复杂工作流时,能实现极速生成。如果你正在构建智能体或多轮对话系统,并追求远超vLLM的推理性能,SGLang是理想选择。
